Patients with diabetes enjoy longer and healthier lives when their blood glucose levels are controlled and their disease is monitored carefully. Experts have long advocated for improved adherence to hypoglycemic medications and secondary prevention measures, especially for underserved populations such as ethnic minorities and the poor. However, a new type of health insurance - high-deductible health plans - may threaten patients' access to crucial medical care. Families in high-deductible health plans must pay up to $4000 per year out-of-pocket before more comprehensive coverage begins. Enrollment in these plans is growing at unprecedented rates. It is essential to understand how high-deductible health plans affect secondary prevention and outcomes among populations with diabetes and in underserved patients. Reduced levels of care could lead to accelerated disease progression and diabetes-related deaths. This proposal seeks to assess the impact of high-deductible health plans on diabetes secondary prevention and outcomes in a nationally representative diabetic population. Measures of secondary prevention will include hemoglobin A1C testing, lipid testing, and retina exams. Outcomes will include high severity emergency department visits, preventable hospitalizations, and hospitalization days. The study will also quantify the degree to which transition to a high-deductible health plan affects socioeconomic and racial/ethnic disparities in diabetes secondary prevention and outcomes. Finally, the study will determine impact of high-deductible health plans with and without full drug cost-sharing on medication adherence and outcomes. This study will draw from a 12-year rolling sample of members from a large national health insurer whose employers mandated a switch from traditional to high-deductible health plans. Employer- and member-level propensity score matching will be used to create a concurrent control group with minimal selection effects. The study will employ a strong quasi-experimental pre-post with comparison group design to examine our outcomes of interest. This will be the first study to examine these research questions on a national scale and the first to use an innovative member-level method for detecting at-risk populations. Results should help inform policy makers, patients, and physicians regarding optimal health plan benefit designs for patients with diabetes.